Hearing difficulties - whether caused by hearing loss or irrelevant ambient noise - can affect more than one's ability to identify speech; when a stimulus is difficult to perceive, memory for that stimulus suffers [47]. Indeed, this memory detriment is present even when the stimulus has been correctly identified [32, 40]. To determine how this effect arises, we examined the role of difficult perception in recall for word-lists [38]. Participants listened to lists where a single word within a list was acoustically masked, and were then asked to recall the lists. The level of masking was adjusted such that the masked word was difficult, but not impossible, to identify. This masking reduced the recall probability for the masked word and, of particular interest, for the non-masked word prior to the masked. These results revealed that difficult listening negatively affects memory for previous, non-degraded speech, as well as for the degraded information itself. Further analysis suggested that this reduction in recall probability was due to a disrupted output pattern. During normal recall, subjects have a strong tendency to follow recall of any word with the word that directly followed it during presentation. This is thought to be due to associations formed between neighboring words during listening. These associations are followed during recall, thereby aiding memory for the list. However, this pattern is disrupted when a masked word is present; there is a reduced probability of following these associations to the masked word, and to words prior to the masked. To provide a mechanism for this effect, we matched the behavioral results to a preliminary computational model. My goal is to formulate an account of associations formed during listening, disrupted by perceptual difficulty, and utilized during recall. I first aim to substantiate the theory that perceptual difficulty alters the formation of associations between words. I will do this by challenging our model against alternative accounts of recall impairment due to poor perception. Second, I will elucidate the structure of the association deficits that come with perceptual difficulty, thereby providing a mechanism for understanding the subtle but important effects of hearing quality on memory for speech. Because the effect of difficult perception has implications beyond list-learning impairment, my final aim is to determine whether association models can account for recall of meaningful materials and passages.